Implemented agent which chooses to play winning, losing or random moves by solving a simplified MDP model of the game using policy iteration.

Portions of MDP/solver code by Ciaran O'Reilly and Ravi Mohan used under MIT license.

src/aima/core/probability/mdp/RewardFunction.java

@@ -0,0 +1,21 @@
+package aima.core.probability.mdp;
+
+/**
+ * An interface for MDP reward functions.
+ * 
+ * @param <S>
+ *            the state type.
+ * @author Ciaran O'Reilly
+ * @author Ravi Mohan
+ */
+public interface RewardFunction<S> {
+	
+	/**
+	 * Get the reward associated with being in state s.
+	 * 
+	 * @param s
+	 *            the state whose award is sought.
+	 * @return the reward associated with being in state s.
+	 */
+	double reward(S s);
+}